Pith. sign in

REVIEW 36 cited by

ToRL: Scaling Tool-Integrated RL

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2503.23383 v1 pith:R2U4FDUR submitted 2025-03-30 cs.CL

ToRL: Scaling Tool-Integrated RL

classification cs.CL
keywords learningmodelsreinforcementtooltool-integratedtorlcomputationalreasoning
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

We introduce ToRL (Tool-Integrated Reinforcement Learning), a framework for training large language models (LLMs) to autonomously use computational tools via reinforcement learning. Unlike supervised fine-tuning, ToRL allows models to explore and discover optimal strategies for tool use. Experiments with Qwen2.5-Math models show significant improvements: ToRL-7B reaches 43.3\% accuracy on AIME~24, surpassing reinforcement learning without tool integration by 14\% and the best existing Tool-Integrated Reasoning (TIR) model by 17\%. Further analysis reveals emergent behaviors such as strategic tool invocation, self-regulation of ineffective code, and dynamic adaptation between computational and analytical reasoning, all arising purely through reward-driven learning.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 36 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Can Agents Generalize to the Open World? Unveiling the Fragility of Static Training in Tool Use

    cs.AI 2026-07 unverdicted novelty 7.0

    Static SFT and RL training for tool-use agents leads to performance drops under open-world distributional shifts across perception, interaction, reasoning and internalization; perturbation-augmented fine-tuning is pro...

  2. SENTINEL: Failure-Driven Reinforcement Learning for Training Tool-Using Language Model Agents

    cs.CL 2026-06 unverdicted novelty 7.0

    SENTINEL generates targeted tasks from model failures in a Controller-Proposer-Solver loop, raising Pass^1 from 66.4 to 74.9 on Tau2-Bench Retail and outperforming standard RL.

  3. Knowing When to Ask: Segment-Level Credit Assignment for LLM Tool Use

    cs.LG 2026-05 unverdicted novelty 7.0

    CARL trains a critic for segment-level credit assignment from binary outcomes in LLM tool-use trajectories, yielding 6.7-9.7 point accuracy gains and 53% fewer calls on solvable questions across five benchmarks.

  4. To Call or Not to Call: A Framework to Assess and Optimize LLM Tool Calling

    cs.AI 2026-05 unverdicted novelty 7.0

    A normative-descriptive framework shows LLMs' tool-calling perceptions misalign with true need/utility for web search, and hidden-state estimators improve decisions over self-perceived baselines.

  5. H$^2$SD: Hybrid Hindsight Self-Distillation

    cs.LG 2026-07 conditional novelty 6.0

    H2SD routes successful RLVR trajectories to token-level magnitude credit assignment and failed trajectories to reverse-KL distillation from a hint-conditioned self-teacher, improving logical reasoning benchmarks.

  6. BVS: Bayesian Visual Search with Multimodal Large Language Model for Fine-grained Perception

    cs.CV 2026-07 conditional novelty 6.0

    BVS combines early-stop attention rollout priors with a scale-aware non-stationary kernel and GP-UCB to locate tiny objects in UHR images more accurately and with fewer MLLM queries than prior visual-search methods.

  7. ECHO: Prune To Act, Trace To Learn With Selective Turn Memory In Agentic RL

    cs.LG 2026-06 conditional novelty 6.0

    Source-indexed selective turn memory plus provenance-guided credit assignment raises BrowseComp-Plus accuracy to 43.4% over GRPO and SUPO while cutting turn proliferation.

  8. ECHO: Prune To Act, Trace To Learn With Selective Turn Memory In Agentic RL

    cs.LG 2026-06 unverdicted novelty 6.0

    ECHO is a selective turn-memory framework for agentic RL that compresses turns into indexed records, selects them for bounded contexts, and uses source indices to assign outcome credit to supporting evidence, reaching...

  9. APPO: Agentic Procedural Policy Optimization

    cs.LG 2026-06 unverdicted novelty 6.0

    APPO refines branching and credit assignment in agentic RL via a Branching Score and procedure-level scaling, improving baselines by nearly 4 points on 13 benchmarks.

  10. Agent Explorative Policy Optimization for Multimodal Agentic Reasoning

    cs.CL 2026-05 unverdicted novelty 6.0

    AXPO addresses the Thinking-Acting Gap in agentic RL training by targeted resampling of tool calls in all-wrong subgroups, delivering +1.8pp gains over GRPO on nine multimodal benchmarks with an 8B model beating a 32B...

  11. Harnessing LLM Agents with Skill Programs

    cs.AI 2026-05 conditional novelty 6.0

    HASP upgrades textual skills into executable Program Functions that intervene in LLM agent loops at inference, post-training, or self-evolution, delivering 25% gains over ReAct and 30.4% over Search-R1 on reasoning be...

  12. ComplexMCP: Evaluation of LLM Agents in Dynamic, Interdependent, and Large-Scale Tool Sandbox

    cs.AI 2026-05 unverdicted novelty 6.0

    ComplexMCP benchmark shows top LLM agents achieve under 60% success on dynamic interdependent tool tasks versus 90% for humans, due to tool retrieval saturation, over-confidence, and strategic defeatism.

  13. ComplexMCP: Evaluation of LLM Agents in Dynamic, Interdependent, and Large-Scale Tool Sandbox

    cs.AI 2026-05 unverdicted novelty 6.0

    ComplexMCP benchmark shows current LLM agents achieve at most 60% success on interdependent tool tasks versus 90% for humans, due to tool retrieval saturation, over-confidence, and strategic defeatism.

  14. PruneTIR: Inference-Time Tool Call Pruning for Effective yet Efficient Tool-Integrated Reasoning

    cs.CL 2026-05 unverdicted novelty 6.0

    PruneTIR prunes erroneous tool-call trajectories during LLM inference via three trigger-based components to raise Pass@1 accuracy and efficiency while shortening context.

  15. SOD: Step-wise On-policy Distillation for Small Language Model Agents

    cs.CL 2026-05 unverdicted novelty 6.0

    SOD reweights on-policy distillation strength step-by-step using divergence to stabilize tool use in small language model agents, yielding up to 20.86% gains and 26.13% on AIME 2025 for a 0.6B model.

  16. To Call or Not to Call: A Framework to Assess and Optimize LLM Tool Calling

    cs.AI 2026-05 unverdicted novelty 6.0

    LLMs often misalign their self-perceived need for tools with true need and utility, but lightweight estimators trained on hidden states can improve tool-calling decisions and task performance across multiple models and tasks.

  17. Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligence

    cs.AI 2026-04 unverdicted novelty 6.0

    Agent-World autonomously synthesizes verifiable real-world tasks and uses continuous self-evolution to train 8B and 14B agents that outperform proprietary models on 23 benchmarks.

  18. When to Trust Tools? Adaptive Tool Trust Calibration For Tool-Integrated Math Reasoning

    cs.CL 2026-04 unverdicted novelty 6.0

    ATTC reduces 'Tool Ignored' errors in tool-integrated reasoning by adaptively trusting tool results according to generated code confidence, yielding 4.1-7.5% gains across models and datasets.

  19. CharTool: Tool-Integrated Visual Reasoning for Chart Understanding

    cs.AI 2026-04 unverdicted novelty 6.0

    CharTool equips MLLMs with cropping and code tools plus agentic RL on DuoChart data to raise chart-reasoning accuracy by up to 9.78 percent on benchmarks.

  20. Multi-Modal Multi-Agent Reinforcement Learning for Radiology Report Generation

    cs.CV 2026-02 unverdicted novelty 6.0

    MARL-Rad trains region-specific and global agents with reinforcement learning on clinical rewards to produce more accurate radiology reports than prior methods on MIMIC-CXR and IU X-ray datasets.

  21. The Reasoning Trap: How Enhancing LLM Reasoning Amplifies Tool Hallucination

    cs.LG 2025-10 conditional novelty 6.0

    Strengthening LLM reasoning through RL, SFT, or chain-of-thought prompting increases tool hallucination rates on SimpleToolHalluBench, with a reliability-capability trade-off observed across mitigation attempts.

  22. The Landscape of Agentic Reinforcement Learning for LLMs: A Survey

    cs.AI 2025-09 accept novelty 6.0

    Survey that defines agentic RL for LLMs via POMDPs, introduces a taxonomy of planning/tool-use/memory/reasoning capabilities and domains, and compiles open environments from over 500 papers.

  23. WebThinker: Empowering Large Reasoning Models with Deep Research Capability

    cs.CL 2025-04 unverdicted novelty 6.0

    WebThinker equips large reasoning models with autonomous web exploration and interleaved reasoning-drafting via a Deep Web Explorer and RL-based DPO training, yielding gains on GPQA, GAIA, and report-generation benchmarks.

  24. ReTool: Reinforcement Learning for Strategic Tool Use in LLMs

    cs.CL 2025-04 unverdicted novelty 6.0

    ReTool uses outcome-driven RL to train 32B LLMs to dynamically use code tools during reasoning, reaching 72.5% accuracy on AIME and surpassing o1-preview.

  25. RSPO: Reward-Swap Policy Optimization for Multi-Turn LLM Agents

    cs.LG 2026-07 conditional novelty 5.5

    Reward-Swap Policy Optimization uses dense-reward exploration to generate diverse trajectories that then train the final policy strictly under true outcome rewards, lifting multi-turn agent performance.

  26. Why Multi-Step Tool-Use Reinforcement Learning Collapses and How Supervisory Signals Fix It

    cs.CL 2026-06 unverdicted novelty 5.0

    RL for LLM multi-step tool use collapses from control token probability spikes but interleaving SFT improves stability at the cost of OOD generalization.

  27. Modularized Reinforcement Learning on LLMs: From MDP Creation to Exploration and Learning

    cs.LG 2026-06 unverdicted novelty 5.0

    Survey mapping RL techniques onto LLM training and highlighting gaps in value-based, off-policy, and bootstrapping methods.

  28. IAPO: Input Attribution-Aware Policy Optimization for Tool Use in Small Multimodal Agents

    cs.LG 2026-06 unverdicted novelty 5.0

    IAPO is an RL method that aligns model input attributions with a teacher to improve tool-calling in multimodal SLMs, reporting 3% average VQA accuracy gains on Qwen2.5-VL-3B across six tests.

  29. Pushing the Limits of LLM Tool Calling via Experiential Knowledge Integration and Activation

    cs.CL 2026-06 unverdicted novelty 5.0

    Instance-level experiential knowledge provides strong gains for LLM tool calling, parallel sampling activates it more effectively than deeper reasoning, and RL-based internalization outperforms SFT, yielding the KATE ...

  30. SlimSearcher: Training Efficiency-Aware Web Agents via Adaptive Reward Gating

    cs.LG 2026-06 unverdicted novelty 5.0

    SlimSearcher reduces tool-call rounds by 17-58% on GAIA, BrowseComp and XBenchDeepSearch while maintaining accuracy via Pareto filtration in SFT and Adaptive Reward Gating in RL.

  31. Tool-Aware Optimization with Entropy Guidance for Efficient Agentic Reinforcement Learning

    cs.LG 2026-06 unverdicted novelty 5.0

    TAO-RL improves agentic RL by filtering degenerate trajectories and reshaping advantages with tool-aware entropy bonuses, yielding better performance on reasoning benchmarks.

  32. E3-TIR: Enhanced Experience Exploitation for Tool-Integrated Reasoning

    cs.AI 2026-04 unverdicted novelty 5.0

    E3-TIR integrates expert prefixes, guided branches, and self-exploration via mix policy optimization to deliver 6% better tool-use performance with under 10% of the usual synthetic data and 1.46x ROI.

  33. Failure Makes the Agent Stronger: Enhancing Accuracy through Structured Reflection for Reliable Tool Interactions

    cs.CV 2025-09 unverdicted novelty 5.0

    Structured reflection makes error diagnosis and repair an explicit trainable step that improves reliability and reduces redundant calls in tool-using LLM agents.

  34. UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning

    cs.AI 2025-09 conditional novelty 5.0

    UI-TARS-2 reaches 88.2 on Online-Mind2Web, 47.5 on OSWorld, 50.6 on WindowsAgentArena, and 73.3 on AndroidWorld while attaining 59.8 mean normalized score on a 15-game suite through multi-turn RL and scalable data generation.

  35. Capability-Aligned Hierarchical Learning for Tool-Augmented LLMs

    cs.AI 2026-06 unverdicted novelty 4.0

    CAHL jointly optimizes hierarchical policies for tool-augmented LLMs via RLVR and reports improved results on API-Bank, BFCL, and Bamboogle.

  36. A Survey of Reinforcement Learning for Large Reasoning Models

    cs.CL 2025-09 accept novelty 3.0

    A survey compiling RL methods, challenges, data resources, and applications for enhancing reasoning in large language models and large reasoning models since DeepSeek-R1.